Remove throughput figures: limited test hardware

README.md

@@ -79,7 +79,7 @@ Mainline llama.cpp does not load `clover_moe`. This build adds the converter, th
 | Windows x86_64 | Any x86-64 CPU | per-CPU DLLs, AVX-512 off | shipped |
 | macOS, ARM64, ROCm on any OS | - | none | not published |
 
-Throughput measured with the NanoPlus GGUF in this build: 6.75 t/s on CPU, 8.51 t/s on CUDA at `-ngl 8`. Vulkan was only smoke-tested on a basic Intel iGPU to confirm it loads and generates; no throughput claim is made for AMD or Intel GPUs. Coverage here means the model runs, not that every backend is benchmarked.
+Every backend in the table was checked for one thing only: that it loads a Clover GGUF and generates. No speed figures are published, because the only test hardware available is a single constrained machine and its numbers would say more about that machine than about your setup. Coverage here means the model runs, not that every backend is benchmarked.
 
 Windows is native. The ZIP carries its CUDA 12.8 libraries and uses the Vulkan loader from the GPU driver; extract it and run `set PATH=%~dp0bin;%PATH%` from the archive root. It ships the same tool set as the Linux archives, including `llama-speculative-simple`, `llama-quantize` and `llama-imatrix`.